Metal-catalyzed hydroxyl radicals are potent mediators of cellular injury, affecting every category of macromolecule, and are central to the oxidative injury hypothesis of Alzheimer disease (AD) pathogenesis. Studies on redox-competent copper and iron indicate that redox activity in AD resides exclusively within the neuronal cytosol and that chelation with deferoxamine, DTPA, or, more recently, iodochlorhydroxyquin, removes this activity. We have also found that while proteins that accumulate in AD possess metal-binding sites, metal-associated cellular redox activity is primarily dependent on metals associated with nucleic acid, specifically cytoplasmic RNA. These findings indicate aberrations in iron homeostasis that, we suspect, arise primarily from heme, since heme oxygenase-1, an enzyme that catalyzes the conversion of heme to iron and biliverdin, is increased in AD, and mitochondria, since mitochondria turnover, mitochondrial DNA, and cytochrome C oxidative activity are all increased in AD. These findings, as well as studies demonstrating a reduction in microtubule density in AD neurons, suggest that mitochondrial dysfunction, acting in concert with cytoskeletal pathology, serves to increase redox-active heavy metals and initiates a cascade of abnormal events culminating in AD pathology.